MCE Erros

[Thomas Glanzmann <thomas@glanzmann.de>]

Hello,
I have two Dual Opteron Machines where I get two MCE errors on. The
first one is:

        MCE 0
        HARDWARE ERROR. This is *NOT* a software problem!
        Please contact your hardware vendor
        CPU 0 4 northbridge TSC edc587de6e99
        ADDR 1001a0000
          Northbridge GART error
               bit61 = error uncorrected
          TLB error 'generic transaction, level generic'
        STATUS a40000000005001b MCGSTATUS 0

I see this error exactly 8 times. What does 'GART' mean?

And here is another one another box:

        MCE 0
        HARDWARE ERROR. This is *NOT* a software problem!
        Please contact your hardware vendor
        CPU 1 4 northbridge TSC f23151075b21d
        ADDR b8898250
          Northbridge Chipkill ECC error
          Chipkill ECC syndrome = f858
               bit32 = err cpu0
               bit46 = corrected ecc error
               bit62 = error overflow (multiple errors)
          bus error 'local node origin, request didn't time out
              generic read mem transaction
              memory access, level generic'
        STATUS d42c4001f8080813 MCGSTATUS 0

How do I identify the broken Memory Modules?

        Thomas